All patients underwent a 'standard' low anterior rectal resection with total mesorectal excision [1,4]. Briefly, surgery was performed under general anesthesia. The sigmoid colon and rectum was removed with high ligation of the inferior mesenteric artery. Autonomic nerves were preserved [5]. The descending colon was anastomosed side-to-end to the anal canal by using the double-stapling technique (CEEA stapling device [Ethicon, Cincinnati, OH, USA] with an outer diameter of 29 mm or 31 mm). A diverting loop ileostomy was routinely performed at a site that had been marked previously by the stoma nurse. All the patients had negative margins on subsequent histological examination. Anorectal function was assessed by using anorectal manometry preoperatively and at 3 months and 4 months after the first operation following closure of the stoma. The Wexner scores [6] of the patients were recorded preoperatively and 4 months from the first operation.

The Wexner score examines the frequency of three types of faecal incontinence (solid, liquid, and gas) and their consequences (pad wearing and lifestyle alteration). For each item, the 5 frequency options range from never (score 0) to always (meaning at least once per day; score 4). The total score is the sum of the item scores and ranges from 0 (perfect continence) to 20 (complete incontinence).

The patient was asked to miss one meal prior to the study and to evacuate his or her rectum at least 2 hours before the measurement was taken. The patient was placed in the left lateral position; then, a perianal inspection and a digital rectal examination were performed, after which the patients was given instructions concerning the anorectal manometry examination. We used a Memphis Biomedica 8-channel water-perfused flexible catheter (Medicalpuls, Bologna, Italy) with side-holes for pressure recording, and the procedure was performed using a station pull-through technique. A computerized system (Polygram Lower GI, Synectics Medical, Stockholm, Sweden) was used for data acquisition. After calibration and exclusion of air from the system, the transducer was placed in the rectum and withdrawn at a rate of 5 mm/sec. The recording was started after a 5-minute period of stabilization. First, the mean basal pressure was recorded; then, the patient was asked to voluntarily squeeze his or her perianal muscles as hard as possible and to retain the contraction for as long as possible. The maximum squeeze pressure and the duration of maximum squeeze were, thus, recorded. A standard balloon-tipped catheter with transducer was then placed with the balloon 8 cm from the anus. The perception of rectal filling and the capacity of the neorectum were measured by placing the balloon with its lower edge 5 cm above the anal verge and subsequently slowly, in a stepwise manner, inflating the balloon with air in 20-mL increments. The thresholds of the patient's first perception of rectal filling, urge to defecate, and maximal tolerable volume were assessed.

Data are presented as means ± standard deviations. Pre- and post-operative serial values were compared by using the Student t-test. Significance was assumed for P < 0.05. The Bland-Altman method was applied to evaluate the agreement between anorectal manometry and the Wexner score [7]. That method calculates the mean difference between two methods of measurement and the 95% limits of agreement as the mean difference (1.96 of the standard deviation of the difference). The 95% limits are expected to include 95% of the differences between the 2 measurement methods. The plot of such data is commonly called a Bland-Altman plot, and the associated method is usually called the Bland-Altman method.